EU Bans Non-Consensual AI Deepfakes

🔑 Enhanced Key Takeaways

• •The ban was included in the final compromise text by the Cyprus Presidency of the EU Council, with explicit support from France, Spain, Germany, and Slovakia[1].
• •The amendment targets AI models likely to generate deepfakes and non-consensual sexually explicit content as part of the digital simplification package[1].
• •Mandatory labeling of AI-generated content, including deepfakes, becomes enforceable from August 2026 under broader AI Act transparency rules[3][5].
• •The European Commission fined X under the DSA in December 2025 for transparency breaches and ordered retention of Grok-related documents amid ongoing investigations[4].

🔮 Future ImplicationsAI analysis grounded in cited sources

⏳ Timeline